Probabilistic Fatigue Life Prediction of Plain Concrete Under Cyclic Loading Through Experimental Finite Element and Stochastic Modelling

The fatigue failure of plain concrete under cyclic loading remains a challenge in structural engineering, particularly for infrastructure exposed to repeated stress fluctuations. Existing fatigue life prediction models fail to fully capture material variability and uncertainties, leading to inconsis...

Full description

Saved in:
Bibliographic Details
Main Authors: Mohamad Shazwan Ahmad Shah, Sarehati Umar, Nurul ‘Azizah Mukhlas, Ng Chiew Teng, Haikhal Faeez Hairuddin, Wan Ikram Wajdee Wan Ahmad Kamal, Hanis Hazirah Arifin, Norhazilan Md. Noor
Format: Article
Language:English
Published: ARQII PUBLICATION 2025-06-01
Series:Applications of Modelling and Simulation
Subjects:
Online Access:https://arqiipubl.com/ojs/index.php/AMS_Journal/article/view/911/223
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The fatigue failure of plain concrete under cyclic loading remains a challenge in structural engineering, particularly for infrastructure exposed to repeated stress fluctuations. Existing fatigue life prediction models fail to fully capture material variability and uncertainties, leading to inconsistent estimations and potential structural failures. This study develops a probabilistic finite element model to predict the fatigue life of plain concrete beams under cyclic loading. Fourteen data points from experimental tests and numerical simulations were analysed, incorporating probabilistic methods to enhance predictive accuracy. Fatigue behaviour was characterized through stress–life (S–N) curves, and regression analysis established the fatigue life equation. The integration of exponential distribution models enabled probabilistic fatigue life estimation across stress levels and failure probabilities. The methodology involves finite element simulation with validated mesh convergence, integration of experimental fatigue data, and application of exponential probability distribution for multi-probability failure predictions. This study directly supports SDG 9 by promoting resilient infrastructure through advanced fatigue life prediction techniques, enhancing the durability and serviceability of concrete structures under cyclic loading. Additionally, mesh convergence validation ensured computational accuracy, identifying an optimal mesh size for reliable numerical simulations. The final logarithmic S–N curve exhibited a high correlation coefficient (R2 = 0.989) between experimental and simulated data. The results underscore the importance of integrating deterministic finite element methods with probabilistic modelling to achieve a robust reliable fatigue life prediction framework for concrete structures. This research advances understanding of fatigue performance in plain concrete, providing essential insights for enhancing infrastructure durability and reliability.
ISSN:2600-8084